Reserve volume estimation of oil storage tanks based on remote sensing images

Pantos, Athanasios; Πάντος, Αθανάσιος

dc.contributor.author	Pantos, Athanasios	en
dc.contributor.author	Πάντος, Αθανάσιος	el
dc.date.accessioned	2022-12-16T11:38:56Z
dc.date.available	2022-12-16T11:38:56Z
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/56490
dc.identifier.uri	http://dx.doi.org/10.26240/heal.ntua.24188
dc.rights	Αναφορά Δημιουργού-Μη Εμπορική Χρήση-Όχι Παράγωγα Έργα 3.0 Ελλάδα	*
dc.rights.uri	http://creativecommons.org/licenses/by-nc-nd/3.0/gr/	*
dc.subject	High-resolution remote sensing image	en
dc.subject	Shadow length calculation	el
dc.subject	Tank volume calculation	en
dc.subject	Object detection	en
dc.subject	Yolov3	en
dc.subject	Shadow extraction	en
dc.title	Reserve volume estimation of oil storage tanks based on remote sensing images	en
heal.type	bachelorThesis
heal.classification	Machine Learning	en
heal.language	en
heal.access	free
heal.recordProvider	ntua	el
heal.publicationDate	2022-07-27
heal.abstract	This work aims to derive oil tank volumes from high-resolution satellite data in order to satisfy the growing demand for accurate measurements of oil tank volumes around the world. Using Otsu thresholding, shadow area thresholding, and morphological closing, the shadow of the oil tank is retrieved from a remote sensing HSV image. The shadow of the oil tank is in the shape of a crescent. The height and the radius of the tank are used in the calculation of the total volume of the tank. In order to assess the viability of the suggested method, an optical remote sensing image with a high resolution was obtained from the Motoroil refinery located in Agioi Theodoroi, Greece. On the test set, the object detection algorithm known as yolov3 was able to get an AP score of 0.84, while on the train set it earned a score of 0.942. In addition, the amount of crude oil barrels that were projected to be contained in the floating head tanks that were discovered was compared to the actual number of tanks. According to the findings, the method that was suggested is effective at calculating the capacity of oil storage tanks to a high degree of precision and with an accuracy that is appropriate for use in actual applications.	en
heal.advisorName	Karantzalos, Konstantinos	en
heal.committeeMemberName	Grammatikopoulos, Lazaros	en
heal.committeeMemberName	Stamatakis, Emmanuel	en
heal.academicPublisher	Εθνικό Μετσόβιο Πολυτεχνείο. Σχολή Ηλεκτρολόγων Μηχανικών και Μηχανικών Υπολογιστών	el
heal.academicPublisherID	ntua
heal.numberOfPages	59 σ.	el
heal.fullTextAvailability	false